<table border="1" cellspacing="0" cellpadding="8">
    <tr>
        <th>Issue</th>
        <td>
            <a href=https://github.com/llvm/llvm-project/issues/106867>106867</a>
        </td>
    </tr>

    <tr>
        <th>Summary</th>
        <td>
            [clang] -Wunused-value does not warn about `literal;` if the literal is expanded from macro defined in system header
        </td>
    </tr>

    <tr>
      <th>Labels</th>
      <td>
            clang
      </td>
    </tr>

    <tr>
      <th>Assignees</th>
      <td>
      </td>
    </tr>

    <tr>
      <th>Reporter</th>
      <td>
          sleeptightAnsiC
      </td>
    </tr>
</table>

<pre>
    Hi,

given following C code:

```C
#include <stdbool.h>
int main (void)
{
    true;
    return 0;
}
```
```
$ gcc -Wunused-value main.c
$ clang -Wunused-value main.c
$ icx -Wunused-value main.c
```

I would expect that `-Wunused-value` would warn about `true;` statement
but it does not - not in GCC 14.2, nor in Clang 18.1, nor in Intel Compiler 2024.1.0
not unlsess I specify `-Wsystem-header`

However,
this functionality was working as expected back in GCC 3.4.6 and Clang 3.0.0
(regression / something has changed? see gobolt link down below)
and recent C++ front-ends DO catch this unlike C front-ends:

```
$ g++ -Wunused-value main.c
main.c: In function ‘int main()’:
main.c:4:9: warning: statement has no effect [-Wunused-value]
    4 |     true;
      |         ^
$ clang++ -Wunused-value main.c
clang++: warning: treating 'c' input as 'c++' when in C++ mode, this behavior is deprecated [-Wdeprecated]
main.c:4:5: warning: expression result unused [-Wunused-value]
    4 |     true;
      |     ^~~~
```

MSVC 19.40 also catches this with C11 Wall flags:

```
> cl /Wall /nologo /std:c11 main.c
main.c
main.c(4): warning C4555: result of expression not used
```

I've checked everything from GCC bugzilla's meta bug about `-Wunused` warnings and found nothing about it.
I also checked issues in LLVM github repo with the same result - nothing found.

I'm a bit concerned that since so many C compilers give the same result, this might be some kind of "agreed between everyone" behavior, but it looks like a bug so I'm reporting it as a bug :)

If it's not-a-bug, then PLEASE, give some rationale on this topic.
If this an actual bug then I can also send it further to GCC.

References:
https://godbolt.org/z/joK6348rE
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77299
https://stackoverflow.com/questions/78935304/why-wunused-value-does-not-catch-a-statement-true/78935413#78935413


</pre>
<img width="1px" height="1px" alt="" src="http://email.email.llvm.org/o/eJysVsGO47gR_Rr6UrAgUZJlH3zoVtvZRmaRIAvsHgOKLElcU6RDUvb0HubbA1Ky7DE6M5cAhkyVisWqx1ePZM7JTiPuSflKyrcVG31v7N4pxLOXXe9ftJP1qjHiY_-LJLQm6RtJX6ZnJy-ooTVKmavUHdTAjUCSvzw6kU06_W5TaS41V6NAIHntvGiMUUlP8sP0XWoPA5MaCN1ejBSE7uaJ1es0AADwdkSSPxgs-tFqSBcjqd6eEvj8lRbQcQ7rP0Y9OhTrC1MjxgwSfnfhiunuZ06Sf_2hy9PK8fkOVzMqAfj1jNyD75kHskm_D0M26ex2ZVYDa8wYvW4obFJwnnkcUPspbDN6kB6EQQfaeFjHp9Twt7qGrEgooTVoY4OpjrVl2yR7ML5rjwpqM5ylQgs0pUWSJXPqIdaolUPn4B3cGblsP6a03YfzOKx7ZALtU62_mCte0C4s8r100I6ae2k0U9J_wJU5uBp7CnRibkYFBTSMn27p50mRbIBpMWeeJ2my7ObWYmfROWkCg47gzIC-D-F65oD3THcoSH4EhwidaYzyoKQ-gTBXDQ0qc10YF5awyFF7qAl9JfQVWmu0X6MWDt7-AZx53kMsY9RKnhDqB4__1QgPxJuj_oA08zh_gXe9YAXkQMk2JbvtrVsI3Ya0Z_NuWXqZXpD8JZgjg6TuwnDhTMRGG8C2DSQk5esT_8q3e6cVQKoaPm1CWD7FcXl4aqCfl_vg9pSst8h82EdCK05oBVKfRx9IMhniFFrBtUcdWT0vNgRJovW0Sw327CIDwx0IPFvkLLArFnx_X8r9Dr3yKSH8er4RzaIbVWiJUNX_Az5SHr59-_YD2fj1t99ryHZJkQJTzkxMRDdVeZW-hzrL4A-mFLSKdT_jYn4ArkK_xBmEHrVRpjNh5Lwg-QvPss9o-Tim2yJQ8A4S1EVZRtRmfEz7CFoUEYfiR-JIaHVB4D3yEwoI4vExNXNrzRDFoBm7v6RSjNDKwYCeBctdIW8bERV0SstF6WjNqEXIIYab_KVPbqI8gTqvK50b0QVSffny-6_QSd-PDVg8mwlq3yM4NuCtzvUSN66SPJc0AINGeuBGc7QaxST7TmqO4AwMTH_Ew3QSXwfhoH1eZeH0EI5paDBKHZykFgFoQinrLAbpRH9F1BN6RiOhdOmDEGQ-KpQxJwdRxCYMnYEp2VCoja0nY79NnwOjlqN5Kq0F6eM-aOPXbN2M3ZQkavjnl8PLb4fwGmuJqVo26T6C0VMp3pwlT5Zo0cY0MO5HpuKqMdg78GANW-RQi5BWO1rfowVvAiu-Q_xf2KJFzfHeBL335_hGj4QeOyPCMZAY2xF6_IvQ45_m75u82NrDp-6cJ50eZ_c7_46uN9d_N2OX8E6S_CgFyd-qiu52n0VxnvGTuaBtlbkm3AyEHv8zoguQOEKP1XaXl3laEHq89h_r66OerMOxvg4gx7Zfs_Wi5OsoKvP0IssJzZfhAyYrsc_FLt-xFe6zipZZnu6qctXvt3mRVjRti12b55t2m26yTS5QYMFSvqP5Su7DVSDd5lm2S2lZJqXIs4xXTZNWIuO7jBQpDkyqRKnLEEBaxfbZZ-lmu6lWijWoXLxuUjqrPQ03T7sPEwJpHClSJZ139xBeehXvqNOM8u35DFkuOt9fkJT0aJma70iyjU00G8MZgF_PTAsUk54MjFsDAlsZelJqmC4zMF1mVqNV-ycqRCWYdy8kO_-tz9b8idwTepy0g9DjXP9lT_8bAAD__6M5gWQ">